Hi All. I have a database that is currently sitting on 760MB with 425K records and from the trend it will grow at about 200K records a year now. as it is primarily used on mobile devices I think that this growth will lead to local processing problems. I am looking for a way to "hive off" any records older than say 18 months in the distributed version of the database while keeping all records in an "archive" version and then be able to add new and modified records monthly to the "archive" and removing the oldest month from the distributed version. There are 27 tables involved. Any pointers for achieving this would be appreciated *

when you say distributed version are you saying it is off line or is the main file hosted?

One solution to my mind is use of MirrorSync to keep a central database of all data hosted and users could sync down only a subset of records. 

 

7 minutes ago, Ocean West said:

when you say distributed version are you saying it is off line or is the main file hosted?

One solution to my mind is use of MirrorSync to keep a central database of all data hosted and users could sync down only a subset of records. 

 

The working files are offline and the main file is hosted. We spend a lot of time without even electricity never mind an internet connection in South Africa so we cannot work with an online file. We do in fact use Mirrorsync. It is not the data being synced that is starting to create problems it is the volume that has to be found and organised while on site. There is also the consideration of expensive bandwidth. If 20 people have to download a new file in the morning and perform a first sync that creates problems on our meagre resources so the idea is to minimise the data that the mobile devices have to both download and process while somehow also keeping a total database with the historical data that we need.

is the distributed file the same as the main file or do you have separate Hub and Spoke files?

 

It is exactly the same and when any development takes place,(at least once a week), a new file must be downloaded to each device and an initial sync has to take place. 

MirrorSync has the option now to perform initial sync on server so that you could restrict the amount of data the file needs.